This is a postprint/accepted manuscript of 1
Aune, Jens B., Coulibaly, Adama & Woumou, Kamkam (2019) Intensification of dryland farming in 2
Mali through mechanisation of sowing, fertiliser application and weeding. In Archives of Agronomy 3
and Soil Science 65(3) pp. 400-4110 4
DOI: 10.1080/03650340.2018.1505042 5
6
Intensification of dryland farming in Mali through mechanisation of sowing, fertiliser 7
application and weeding 8
9
Jens B. Aunea, Adama Coulibalyb and Kamkam Woumoub 10
11
aDepartment of International Environment and Development Studies, Norwegian University 12
of Life Sciences, Ås, Norway; bInstitute d’Economie Rural, Bamako, Mali, 13
14
CONTACT: Jens B. Aune, Department of International Environment and Development 15
Studies, Norwegian University of Life Sciences, Ås, Norway; Email: [email protected] 16
orcid.org/0000-0001-5270-1798 17
18
Abstract 19
This study focuses on the role of mechanised sowing and weeding in combination with seed 20
priming and fertiliser microdosing in Mali. Mechanised sowing and weeding were based on 21
using a combined donkey-drawn planter/weeder and a motorised planter/weeder. The research 22
methods included studies of seed delivery in manual and mechanised sowing, field 23
experiments on different levels of mechanization/intensification, labour studies on 24
mechanisation and an economic assessment of the different levels of intensification.
25
The average sorghum grain yield across three years increased by 352 kg ha-1 (43.7% increase) 26
by combining mechanisation with seed priming and microdosing of 0.2 g NPK 15-15-15 27
fertiliser per pocket compared to a control with manual sowing but without seed priming and 28
microdosing. The labour demand (sowing and weeding) for manual, donkey-drawn and 29
motorised operations was respectively 184, 67 and 47 hours ha-1, respectively.
30
An economic analysis showed that the donkey-drawn planter/weeder is the appropriate 31
mechanisation below six ha while above this land size it becomes increasingly interesting for 32
the farmers to invest in a motorised planter. The use of mechanisation will result in earlier and 33
uniform crop establishment, facilitate microdosing application, timelier weeding, higher 34
yields, better economic return and reduced labour demand.
35 36
Keywords: sorghum, seed priming, microdosing, planter, weeder, labour-use, appropriate 37
mechanisation 38
39
Introduction 40
The sequence of agricultural mechanisation goes usually from manual labour, through animal 41
traction to the use of combustions engines (tractors). As annual cropping is developed, the 42
necessity for mechanisation increases due to higher labour demand for tillage, fertilization and 43
weeding (Pingali et al. 1987). The farmers in the drylands of West Africa are generally 44
subsistence orientated and the surplus generated is very limited making savings difficult. This 45
makes it challenging for farmers to invest in mechanisation and purchase of agricultural 46
inputs. Additional constraints for adaptation of mechanisation include low prices for 47
agricultural produce and high prices for agricultural input.
48
Despite some success stories with mechanisation particularly in relation to cultivation of cash 49
crops, its introduction has encountered difficulties in the drylands of West Africa. The 50
profitability of mechanisation increases if it can be used in multiple operations such as tillage, 51
planting, weeding, threshing and transport (Williams 1997). In marginal environments it is 52
suggested that animal traction should be used for the transportation of water, manure and farm 53
produce (Williams 1997). In areas where the cropping season is short and the soil is sandy, 54
the use of planters has been introduced without prior ploughing (Pingali et al. 1987).
55
In general, agricultural mechanisation in Africa has not given satisfactory results (Fonteh 56
2011). Definitions of agricultural intensification in Africa also often overlook the importance 57
of mechanisation. An example is the Montpellier Panel Report (2014) which defines 58
agricultural intensification as a combination of ecological, genetic and socio-economic 59
processes while completely leaving out mechanisation from the intensification definition.
60
Another recent definition states that agricultural intensification is about producing more yield, 61
increasing the number of crops per year or cultivating more high-value crops (Pretty and 62
Bharucha 2014).
63
Agricultural mechanisation has not been on the agenda of most development agencies since 64
1985 (Mrema 2011). However, there is currently a renewed interest in agricultural 65
mechanisation among development actors such as the African Development Bank (AfDB 66
2016) and FAO (FAO 2016). The outlook for agricultural mechanisation may also have 67
changed, as conditions have recently become more favourable for agricultural intensification 68
than in the previous decades. Food prices are on the increase, wages are higher and young 69
people go to the cities in search of jobs (Mrema 2011, Baudron et al. 2015). Mechanisation 70
may make agriculture, which is currently associated with drudgery (Leavey and Hossain 71
2014), more attractive to young people. Appropriate mechanisation can reduce labour demand 72
in peak periods and thereby even out the labour demand throughout the season.
73
Climate change also contributes to an increased need for mechanisation in Africa. The aridity 74
of the climate is expected to increase in the Sahel due to increasing temperatures (Sylla et al.
75
2016) making it necessary for farm operations to be done faster as the time window for these 76
operations will get shorter.
77
In Mali, agricultural mechanisation has been promoted by the parastatal Compagnie Malien de 78
Développement de Textile (CMDT) (Ashburner and Kienzle 2011). CMDT provided credit for 79
agricultural inputs and loans for agricultural machinery that were repayable over several 80
seasons. Planters were first introduced to Senegal in the 1920s and 1930s (Pingali et al. 1987) 81
and later to Mali. The national factory Societé Malienne d'Etudes et de Construction de Matériel 82
Agricole (SMECMA) was established in the 1970s in Mali for the production of planters.
83
However, SMECMA was not able to survive the harsh economic and political conditions of the 84
1980s and 1990s, and the factory was closed down due to supply problems, great variation in 85
demand and organizational problems (Le Thiec and Havard 1996). In addition, CMDT faced 86
problems resulting from structural adjustment policies and support for mechanisation were 87
abandoned. Despite this, the national factory has contributed greatly to mechanisation in Mali.
88
In the 1990s, it was assessed that 70% of farm households were equipped with animal traction 89
in southern Mali. However, only 17% of farmers cultivating the dryland cereal crops (sorghum 90
and millet) had mechanised equipment (DNGR 2005). Manual sowing of millet and sorghum 91
is a demanding operation that includes opening a small pocket in the soil with a thin-bladed hoe, 92
taking a pinch of seeds, placing the seeds in the pocket and covering the seed with the foot. In 93
recent times, farmers are increasingly combining sowing with the use of fertiliser microdosing.
94
Local blacksmiths were trained by CMDT in the 1980s on the construction and maintenance 95
of the planter. Since the demise of SMECMA, these blacksmiths have been ensuring the 96
supply of planters in Mali. The blacksmiths can produce the equipment at 30-50% of the price 97
of larger industrial producers (Pingali et al. 1987).
98
The central hypothesis in this paper is that agricultural intensification based on mechanised 99
sowing and weeding in combination with seed priming and fertiliser microdosing is a feasible 100
option for farmers in West Africa. The paper shows the effect of intensification on yield, 101
labour use, investment needs and economic return. The sustainability of different levels of 102
intensification is also discussed.
103 104
Materials and methods 105
The methods used in this study include studies on seed and fertiliser delivery in manual and 106
mechanised sowing (1), field experiments to determine yields in treatments (2), time and fuel 107
use studies of these treatments (3), and an economic assessment to compare the different 108
levels of intensification and mechanisation options (4). These methods in combination with a 109
national census were used to assess the feasibility of the different mechanisation options 110
taking into consideration labour availability in the household and farm size.
111
Seed delivery in manual sowing and by planter 112
Seed and fertiliser application in manual sowing were based on measuring the quantity of 113
seeds and fertiliser in a pinch between the thumb and the index finger of 75 different farmers.
114
The quantities applied with the pinch were taken with different pearl millet and sorghum 115
varieties, NPK (15-15-15) and diammonium phosphate fertiliser (DAP 18-46-0) and a 1:1 116
mixture of seeds and fertiliser. Assessment of the numbers of seeds applied when using the 117
SMECMA planter was based on running the planter 61 times (corresponding to a row length 118
of 390 m) for each treatment. The essential parts of the SMECMA planter consists of the 119
hopper, the rotating disc with perforations that deliver seeds in the correct quantity and at 120
appropriate spacing, seed delivery tubes, the furrow opener, tines that close the furrow, and a 121
compaction wheel that compresses the soil to increase the contact between soil and seeds.
122
Factors influencing the amount of seeds and fertiliser applied are distance between the 123
perforations in the disc, the diameter of the perforations and the thickness of the disc.
124
Agronomic trials 125
One series of field experiments was conducted in two villages in the Koulikoro region during 126
2007 and 2008 to determine the yield effect of manual sowing compared to mechanised 127
sowing using a donkey-drawn planter. In each village, 10 farmers hosted the test and the plot 128
size for each treatment was 1000 m2. Each farmer represented a replicate.
129
Another series of field experiments was conducted from 2013 to 2015 in Koulikoro region to 130
assess the yield performance at different levels of intensification including mechanization.
131
The trial was conducted in the fields of 13 farmers and each farmer was considered as a 132
replicate. The following treatments in sorghum were used:
133
1. Manual sowing without seed priming or fertiliser 134
2. Mechanised operations using donkey-drawn planter/weeder, and seed priming for 8 135
hours followed by 2 hours surface drying to reduce the stickiness of the seeds. The 136
disc used in the planter had 7 mm diameter perforations and the disc thickness was 8 137
mm.
138
3. Mechanised operations using donkey-drawn planter/weeder, seed priming and 0.2 g 139
NPK per pocket (5 kg NPK ha-1). Seed priming and drying as in treatment 2, followed 140
by mixing seeds and fertiliser at 1:1 volumetric ratio. The disc in the planter had 141
perforations of 10 mm diameter and a disc thickness of 8 mm.
142
Labour assessment 143
Time use was measured for the treatments of manual sowing (1), use of a donkey-drawn 144
planter (2) and motorised planter (3) in 2013, 2015 and 2016. The treatments were replicated 145
in the fields of seven farmers and the plot size for each treatment was 1000 m2. In the case of 146
interruptions of the work, the chronometer was stopped. In the treatment with motorised 147
sowing, the fuel use was measured by first emptying the tank of the planter, thereafter filling a 148
measured quantity of fuel and measuring the remaining fuel again after the operation.
149
The calculation of labour use for sowing per farm was calculated based on the average farm 150
size in Mali, the number of active labourers per farm and the labour demand per ha for the 151
three different sowing methods.
152
Economic assessment 153
In the economic assessment, we used the yield from the experiments conducted from 2013 to 154
2015. We also introduced a fourth level of intensification that is based on the use of a 155
motorised planter in combination with seed priming and fertiliser microdosing (video). This is 156
the same planter that was used at the third intensification level, but in this case the planter is 157
not pulled by traction animals, but by a 5.2 KW (6.8 horsepower) combustion engine (small 158
motorcycle engine). The motorised planter is constructed by “Agric Construction Cissé et 159
Frerès” in Koutiala Mali. We do not have the yield data for the fourth step in the 160
intensification ladder, but the same yield data as in the third level of intensification was used 161
because the seed delivery system is the same as in the planter drawn by traction animals.
162
The economic assessment of the different levels of mechanisation was undertaken using the 163
method described by Sims and Kienzle (2015). This method takes into consideration the 164
depreciating value of the machine, useful life, interest costs, repair costs and the cost of 165
operating the machine. The useful life of the machine was set at 10 years and the interest rate 166
was 12%, a rate typically used in small-scale agricultural credit schemes in Mali. The annual 167
depreciation cost was set to 9% of the price of the machine while the annual repair cost was 168
set to 13% of the sales price of the machine. The sorghum grain price used was 119 CFA- 169
Franc kg-1, which is the average grain price for 2015 across cereal growing regions in Mali.
170
The price of the straw was set to 20 CFA-Franc kg-1 (obtained from a survey). The time for 171
manual weeding in the control was set to 120 hours ha-1, which is the average time for manual 172
weeding estimated in Mali, Burkina Faso and Niger (Memento de l’Agronome 2009). Manual 173
weeding within the row for the mechanised treatment was set to 36 hours ha-1 (Memento de 174
l’Agronome 2009). A survey among 29 farmers showed that the average price for renting a 175
donkey and hiring a man is 3200 CFA-Franc day-1. To calculate the farm partial income for 176
different farm sizes, we calculated the value of the straw and grain yield for farm sizes 177
varying from 1 to 12 ha and subtracted the variable cost related to sowing and weeding for the 178
corresponding farm size and then subtracted the fixed cost related to mechanization.
179 180
Results 181
Manual and mechanised seed and fertiliser application 182
The seeding rate in manually application and by the planter were assessed. One pinch of seeds 183
taken between the thumb and the index finger (farmers practice) gave 35 (standard deviation 184
17) and 11 (standard deviation 5) seeds respectively for the Toroniou millet variety and the 185
CSM sorghum variety (Table 1). When Toroniou seeds and NPK fertiliser (15-15-15) were 186
mixed, one pinch equated to 20 seeds and 0.28 g fertiliser. DAP fertiliser was also tested, and 187
the rate applied was similar to that of NPK fertilizer.
188
Insert Table 1.
189
In mechanised sowing, the application of seeds and fertiliser is determined by distance 190
between perforations in the disc, the diameter of the perforations and the thickness of the disc.
191
Table 2 shows the relationship between the diameter of perforation and the number of seeds 192
delivered for the Toroniou pearl millet variety. The number of seeds delivered increased with 193
3.5 seeds for every mm increase in the perforation diameter. The disc recommended for 194
sowing pearl is a disc with perforations of 8 mm diameter and a thickness of 8 mm, as this 195
disc gave an appropriate number of seeds.
196
Insert Table 2 197
The number of seeds and quantity of fertiliser applied was determined when seeds and 198
fertiliser were mixed in a 1:1 ratio and applied by the planter. This disc had an 8 mm 199
thickness and perforations with a diameter of 10 mm delivering approximately 10 seeds of 200
sorghum or millet and 0.2 g of fertiliser per planting pocket.
201
Use of the planter gave a more uniform sowing rate. The standard deviation for the number of 202
seeds delivered was 16.9 for the manual pinch and whereas as it is 6.8 for the planter. The 203
disc with 13 mm perforations delivered a number Toroniou millet seeds equivalent to a pinch 204
of seeds (Tables 1 and 2).
205
Agronomic effects 206
Mechanisation increased sorghum yield by an average of 14.6% in 2007 (p<0.05) and by 207
13.0% in 2008 (p<0.01). In the trail with three levels of intensification (2013-2015), the 208
average grain yields were 804, 1058 and 1156 kg ha-1 in the treatments manual sowing 209
(without priming and microdosing (1), mechanised sowing and seed priming (2) and 210
mechanised sowing, seed priming and microdosing (3), respectively (Figure 1, Figure 2). The 211
boxplot also showed that there were no yields below 600 kg ha-1 when the highest level of 212
intensification is used.
213
Insert figure 1 and 2.
214
Labour assessment 215
Mechanisation reduced the time used for sowing and weeding. Table 3 shows that the labour 216
demand in sowing and weeding decreased from 184 hours ha-1 in manual sowing, to 67 hours 217
ha-1 when using donkey-drawn traction (Supplemental Figure 1) and to 47 hours ha-1 when 218
using motorised traction (Supplemental Figure 2). The labour demand was therefore 3.9 times 219
higher for manual sowing and weeding compared to motorised sowing and weeding. Labour 220
demand related to sowing is particularly reduced. The reason for this is that even if 221
mechanical weeding is practiced, there is still a need for manual weeding within rows.
222
Insert Table 3 223
Economic assessment 224
Table 4 shows the major fixed and variable cost items related to the different treatments. It 225
appears that the manual treatment has lower fixed costs (independent of area cultivated) while 226
the variable cost per ha is higher for the manual treatment compared with the mechanised 227
treatments. The price of the donkey-drawn planter was 70,000 CFA-Franc (106 Euro) while 228
for the motorised planter the cost was 525,000 CFA-Franc (800 Euro). The major fixed costs 229
for the donkey-drawn planter and motorised planter are connected to depreciation of the 230
machines, interests and repair costs. The fixed costs for the donkey-drawn planter/weeder was 231
20,020 CFA-Franc while it was 150,150 CFA-Franc for the motorised planter/weeder. The 232
variable cost items differed between the treatments. As Table 4 shows, the variable cost 233
decreased from 36,800 CFA-Franc ha-1 in the manual treatment to 13,145 CFA-Franc ha-1 in 234
the motorised treatment. The major reason for this was the higher labour costs in the manual 235
treatment. Mechanised weeding reduced the weeding time to half of that for manual weeding 236
(Table 3). The cost of donkey rental was quite low for the treatments using animal traction.
237
The fertiliser and fuel costs were also low in comparison to the other costs. The amount of 238
petrol consumed per was 3.5 l ha-1 (standard error= 0.03) for sowing and weeding, which was 239
equivalent to 2,625 CFA-Franc.
240
Insert Table 4 241
The partial farm net income (including only the variables investigated in this study) was 242
calculated for cultivated areas ranking from one to twelve ha in order to assess appropriate 243
mechanisation for different farm sizes (Table 5). The data used to calculate the partial income 244
at the farm level were taken from Figure 1, Table 3 and Table 4. The data showed that even if 245
the farmers were cultivating only one ha, it was more profitable to use mechanised sowing, 246
priming and microdosing than to use manual cultivation without seed priming and 247
microdosing. Furthermore, it was shown that if the farmer is cultivating between one and six 248
ha, it was less profitable to use the motorised planter than the other treatments. When farmers 249
cultivate six ha, the partial income in motorised mechanisation and donkey-drawn 250
mechanisation were almost equal (1.8% higher in donkey-drawn mechanisation). Above six 251
ha, the partial farm net income was higher using the motorised planter, compared with the use 252
of the donkey-drawn planter.
253
Insert Table 5 254
255
Discussion 256
Manual and mechanised seed and fertiliser application 257
The higher variability observed in seed and fertiliser delivery in manual application as 258
compared with mechanised application was related to the size of the fingers of the person 259
taking the pinch and how the pinch was taken. In addition, use of the planter gave a more 260
uniform planting distance and sowing depth.
261
Agronomic effects 262
More uniform sowing may explain why the use of the planter gave 14% higher yield than 263
manual sowing. The trial with increasing levels of intensification showed that the highest 264
level of intensification (use of the planter, seed priming and microdosing) increased yield by 265
43.8% compared to farmers’ practice. Seed priming has previously been found to increase 266
yield by about 20-30% compared to farmers’ practices under Sahelian conditions (Aune et al.
267
2017). The seed priming effect was related to a more uniform plant stand and faster crop 268
establishment. Seed priming combined with microdosing has previously been found to 269
increase yield by 106% compared to farmers practice under Sahelian conditions (Aune et al.
270
2012), and this is clearly higher than the yield effects observed in the experiments running 271
from 2013 to 2015.
272
Labour assessment 273
The labour study showed that labour use was 3.9 times higher in manual sowing and weeding 274
as compared to using the motorized planter/weeder, and 2.7 times higher than using the 275
donkey-drawn planter/weeder. Speed of sowing is particularly important in the Sahel as there 276
are few days appropriate for sowing. In order to assess labour availability at sowing for a 277
typical Malian farm, we used data from the national census of farm households in Mali that 278
showed that the average planted areas per farm is 4.7 ha (Direction National d’Agriculture 279
2007). A typical farm household in Mali with four available workers can sow the farm 280
manually in 9.4 days compared to 4.2 and 1.8 days for use of the donkey-drawn and the 281
motorized planter, respectively. This shows that manual sowing will, in many cases, lead to 282
sub-optimal sowing time while the donkey-drawn planter and particularly the motorized 283
planter can ensure timely sowing. The high capacity of the motorized planter may also allow 284
for leasing the planter to other farmers. The lesson from Asia is that small-scale 285
mechanisation has mainly spread through service delivery by owners of 2-wheels tractors (2 286
WT) (Mottaleb et al. 2016, Baudron et al. 2015) and service delivery is also likely to be the 287
most efficient way for promoting mechanisation for African small-scale farmers (Baudron et 288
al. 2015).
289
Economic assessment 290
The economic return to mechanisation will depend on machine costs (depreciation), area 291
cultivated, running cost and yield level. Even if a farmer was only planting one ha, it was 292
better to use the donkey-drawn planter/weeder in combination with seed priming and 293
microdosing than to use manual sowing without seed priming and microdosing. The reason 294
was that manual sowing without seed priming and microdosing will had a labour costs of 295
36,800 CFA-Franc ha-1 while the combined costs of donkey hire and labour in the treatment 296
with the use a donkey drawn planter was 18,720 CFA-Franc ha-1 (Table 4). In addition, the 297
yield was 14% higher with mechanised sowing compared to manual sowing. The cost of 298
fertiliser was very low compared to the labour costs. The benefit of mechanised sowing and 299
weeding increased with increasing planted area as shown in Table 5. The average planted area 300
per farm in Mali is 4.7 ha, making mechanised sowing/weeding an attractive option for the 301
larger farms. It was shown that if farmers cultivate less than six ha it is advisable to use 302
animal traction combined with the yield enhancing technologies, compared to the use of the 303
motorised planer/weeder. Beyond six ha, farmers may choose the donkey-drawn 304
planter/weeder or the motorised planter/weeder combined with the yield enhancing 305
technologies. However, as the farm size increases it becomes more and more difficult to use 306
donkey-drawn mechanisation, because this form of mechanisation does not have the same 307
capacity as motorised mechanisation. For a farm size of six ha, the partial income increased 308
by about 60% when using the donkey-drawn planter combined with the yield-enhancing 309
technologies compared to manual farm operation without any use of the yield enhancing 310
technologies. The exact threshold level at which it becomes interesting to use motorised 311
mechanisation is difficult to determine as there are uncertainties related to depreciation and 312
repair costs when the machines are used more intensively.
313
For some farmers it might be interesting to skip the animal traction stage (stages 2 and 3) and 314
adopt motorised mechanisation since there are many hidden costs in relation to traction 315
animals. These costs are difficult to quantify, but represent significant costs for the farmers in 316
terms of veterinary services, fodder and labour for feeding, herding and training the traction 317
animal. Supplementary feeding at the start of the working period is often needed as the local 318
feed resources are of low quality at this time of the year. A pair of oxen cannot work more 319
than six hours/day and an ox can on average, only deliver traction services for three years 320
(Cattin 1986).
321
Sustainability of intensification- overall assessment 322
Intensification of agriculture in the Sahel has been described as “climbing a ladder or a 323
stairway” (Aune and Bationo 2008). This ladder was based on a stepwise introduction of seed 324
priming, organic fertiliser, microdosing and agroforestry (Figure 3). However, the problem 325
with this ladder is that labour demand increases as new yield enhancing technologies are 326
added to the ladder. Here, we suggest an alternative pathway characterized by combining 327
mechanisation with yield enhancing technologies (seed priming and microdosing) that are 328
compatible with mechanised sowing (Supplemental Figure 2). The steps in the revised ladder 329
were arranged according to increasing costs as in the previous ladder. Donkey-drawn traction 330
and motorized mechanisation represent the second highest and the highest levels of 331
intensification, respectively in the revised ladder. Farmers may choose any step on the ladder 332
depending on their resources and priorities. By climbing the ladder, farming becomes more 333
attractive, particularly for young people, as labour demand is decreased and yield is increased.
334
A combination of mechanisation and yield-enhancing technologies is a well-proven pathway 335
of intensification. The early stages of agricultural intensification in developed countries were 336
also characterized by crop livestock integration, use of farm-yard manure, the introduction of 337
legumes, grazing management and mechanisation (Vos and Meekes 1999, Pretty and 338
Bharucha 2014). Intensification of farming should not increase the probability of crop failure 339
and farmers economic risk. Figure 1 shows that the risk of a low yield is higher in the 340
treatment with farmers practice compared to the treatment with mechanization, seed priming 341
and microdosing. Motorized mechanisation represents a rather high financial cost for the 342
farmer, and many farmers will be in need of credit financing for purchasing a motorized 343
planter.
344
Insert figure 3 345
Motorised mechanisation can be criticised because it will increase CO2 emission. However, 346
these emissions are modest as the total fuel consumption for sowing and weeding was 3.5 lha- 347
1 corresponding to 8 kg CO2 ha-1. The amount of CO2 released for sowing and weeding an 348
average farm of 4.7 ha was therefore about 38 kg CO2. It is also important to keep in mind 349
that there will also be GHG emission if traction animals are used.
350
There is a possibility for using imported 2WT and attachment like planters to promote 351
mechanisation in West Africa, but the advantage with the motorised planter developed by IER 352
is that it can be produced and maintained locally and that the seed delivery system is fine- 353
tuned to deliver seeds and fertiliser at appropriate spacing and quantity. Furthermore, the 354
motorized planter is built on a planter that is well known in Mali.
355 356
Conclusion 357
The suggested intensification pathways based on using mechanised sowing and weeding in 358
combination with the yield enhancing technologies of seed priming and microdosing have 359
clear benefits for the farmer in terms of higher yields, more timely sowing, increased 360
profitability, the saving of labour and reduced drudgery. This intensification pathway 361
therefore increases both land and labour productivity thereby increasing the attractiveness of 362
intensification. This central hypothesis is thus confirmed. The appropriate level of 363
intensification depends on the yield obtained, farm size, labour force of the household, prices 364
of input and output, interest rates and availability of capital. The use of the donkey-drawn 365
planter/weeder combined with seed priming and microdosing seems to be an appropriate level 366
of intensification for farms under six ha while for farmers with land size beyond six ha, the 367
use the donkey-drawn planter or the motorised planters are feasible options. For a farm size of 368
six ha, the partial income will increase by about 60% when using the donkey-drawn planter 369
combined with the yield-enhancing technologies, compared to manual farm operations 370
without any use of yield enhancing technologies.
371 372
Disclosure statement 373
There is no conflict of interest in this study.
374 375
Funding 376
The research was funded by the Norwegian Ministry of Foreign Affairs and the Dryland 377
Coordination Group.
378 379
References 380
AfDB. 2016. African countries urged to prioritize mechanized agriculture for increased 381
productivity. African Development Bank Group, [assessed 2017 Nov 14].
382
https://www.afdb.org/en/news-and-events/african-countries-urged-to-prioritize- 383
mechanized-agriculture-for-increased-productivity-16096/.
384
Ashburner JE, Kienzle J. 2011. Review of public-private sector models for mechanization. In:
385
Ashburner JE, Kienzle J, editors. Investments in agricultural mechanization in Africa.
386
Agricultural and food engineering technical report 8, Rome: FAO, p. 51-58.
387
Aune JB, Bationo A. 2008. Agricultural intensification in the Sahel - The ladder approach.
388
Agr Syst. 98(2):119-125.
389
Aune, JB, Traoré CO, Mamadou S. 2012. Low-cost technologies for improved productivity of 390
dryland farming in Mali. Outlook Agr. 41(2):103-108.
391
Aune, JB, Coulibaly A, Giller KE. 2017. Precision farming for increased land and labour 392
productivity in semi-arid West Africa. A review. Agron Sustain Dev. 37(3):16 393
Baudron F, Sims B, Justice S, Kahan DG, Rose R, Mkomwa S, Kaumbutho P, Sariah J, 394
Nazare R, Moges G, Gerard B. 2015. Re-examining appropriate mechanization in 395
Eastern and Southern Africa: two-wheel tractors, conservation agriculture, and private 396
sector involvement. Food Sci. 7(4):889-904.
397
Cattin MB. 1986. Recherche et dévelopment agricole. Les unités expérimentales du Sénégal.
398
Une publication de ISRA, CIRAD et FAD, 498 p.
399
Direction National de l’Agriculture. 2007. Recensement général de l’agriculture (RGA).
400
Campagne Agricole 2004-2006. [accessed 2017 Nov 14].
401
http://harvestchoice.org/publications/mali-recensement-general-de-lagriculture-rga- 402
campagne-agricole-2004-2005-resultats-defi.
403
DNGR. 2005. Promotion de la mécanisation agricole. Consultation sectorielle sur le 404
développement rural et l’agriculture au Mali. Thème III, Direction National Génie Rural 405
2005, Ministère de l’agriculture, République du Mali.
406
FAO. 2016. Agricultural mechanization. A key input for sub-Saharan African smallholders, 407
FAO, Rome. [accessed 2017 Nov 14]. http://www.fao.org/3/a-i6044e.pdf.
408
Fonteh MF. 2011. Direct public sector investment and financial support to agricultural 409
mechanization in Africa: examples from Ghana and Mali. In: Ashburner JE, Kienzle J, 410
editors. Investment in agricultural mechanisation in Africa, Agricultural and food 411
engineering technical report no. 8, p. 42-46.
412
Leavy J, Hossain N. 2014. Who wants to farm? Young aspirations, opportunities and rising 413
food prices. IDS Working Paper no. 439, 44 p.
414
Le Thiec G, Havard M. 1996. Les enjeux du marché pour la traction animale en Afrique de 415
l’ouest. Agri. Dével. no. 11 : 39-47.
416
Memento de l’agronome. 2009. CIRAD-GRET, Ministère des Affaires Etranger, France, 1691 417
p.
418
Montpellier Panel Report. 2013. Sustainable Intensification: A New Paradigm for African 419
Agriculture, London. [accessed 2016 Nov 20]. https://goo.gl/yUI2bh.
420
Mottaleb KA, Krupnik TJ, Erenstein O. 2016. Factors associated with small-scale agricultural 421
machinery adoption in Bangladesh: Census findings. J. Rural Stud. 46: 155-168.
422
Mrema GC. 2011. Public sector strategy development. In: Ashburner JE, Kienzle J, editors 423
Investments in agricultural mechanization in Africa. Agricultural and food engineering 424
technical report 8. FAO, Rome, p. 23-26.
425
Pingali P, Bigot Y, Binswanger HP. 1987. Agricultural mechanization and the evaluation of 426
farming systems in sub-Saharan Africa. The John Hopkins University Press, 216 p.
427
Pretty J, Bharucha ZP. 2014. Sustainable intensification in agricultural systems. Ann Bot- 428
London 114(8):1571-1596.
429
Sims B, Kienzle J. 2015. Mechanization of conservation agriculture for smallholders: Issues 430
and options for sustainable intensification. Environment 2(2):139-166.
431
Sylla MB, Elguindi N, Giorgi F, Wisser D. 2016. Projected robust shift of climate zones over 432
West Africa in response to anthropogenic climate change for the late 21st century.
433
Climatic Change 134(1-2): 241-253.
434
Vos W, Meekes H. 1999. Trends in European cultural landscape 435
development:perspectives for a sustainable future. Landscape Urban Plan 46(1-3):3- 436
14.on 437
Williams TO. 1997. Problems and prospects in the utilization of animal traction in semi-arid 438
west Africa: evidences from Niger. Soil Till. Res. 42(4):295-311.: N 439
440
Figure 1. Effect of different levels of intensification on stover yield in sorghum 441
442
Figure 2. Effect of different levels of intensification on grain yield in sorghum 443
444
Figure 3. Intensification ladder without mechanization (left) and revised intensification ladder 445
including mechanization (right) 446
447
Table 1. The amount of seeds and gram fertilizer applied by taking a pinch between the thumb and 448
the index finger 449
Pinch with seed or fertilizer Pinch with seed and fertilizer
Toroniou millet (nbr.)
CSM63
Sorghum
(nbr)
NPK fertilizer
g
Toroniou grains
(nbr.)
NPK fertilizer
g
Mean 35 11 0.37 20 0.28
Standard deviation
16.8 4.6 0.21 10 0.17
450 451
Table 2. Relationship between diameter of perforations in the disc and quantity of seed delivered of 452
the Torounio pearl millet variety.
453
8 mm 10 mm 12 mm 13 mm
Number of seeds 16.1 23.4 29.7 32.9
Standard deviation 4.0 5.4 6.3 6.8
454 455
Table 3. Labour demand in h ha-1 for sowing and weeding in the treatments. Standard error for time 456
use in sowing in parenthesis.
457
Manual sowing and weeding
Donkey drawn planter+
seed priming
Donkey drawn planter+ seed priming+microdosing
Motorized planter+ seed priming + microdosing
Labour use per hectare sowing hours
64 (4.3) 7.1 (0.9) 7.1 (0.9) 3.1 (1.2)
Labour use per hectare for weeding
120 601 601 442
Total labour use sowing and weeding per hectare
184 67.1 67.1 47.1
1 Includes 24 hours mechanized weeding between rows and 36 hours manual weeding within rows 458
2 Includes 8 hours mechanized weeding between rows and 36 hours manual weeding within rows 459
460 461
Table 4. Fixed and variable cost for using manual planting, donkey drawn planter and motorized 462
planter in CFA-Franc (1 Euro=656 CFA-Franc).
463
Fixed costs Manual
sowing
Donkey drawn planter+
seed priming
Donkey drawn planter,
seed priming and microdosing
Motorized planter, seed
priming, and microdosing
Depreciations costs 0 6,300 6,300 47,250
Interests costs 0 4,620 4,620 34,650
Repairs costs 0 9,100 9,100 68,250
Total fixed costs per year 20,020 20,020 150,150
Variable costs per hectare
Fertilizer costs 0 0 1,100 1,100
Fuel costs 0 0 0 2,625
Donkey rental cost 0 11,520 11,520 0
Labour costs 36,800 7,200 7,200 9,420
Total variable costs per hectare 36,800 18,720 19,820 13,145
464 465
466
Table 5. Effect of level of intensification and area cultivated on partial farm income in sorghum in 467
1000 CFA-Franc.
468 469
ha Manual Donkey drawn
planter + priming
Donkey drawn planter + priming
+ microdosing
Motorized sowing + priming
+ microdosing
1 119 163 175 62
2 239 345 369 276
4 472 710 759 702
6 716 1,075 1,148 1,128
8 954 1,440 1,537 1,555
10 1,193 1,805 1,927 1,981
12 1,431 2,170 2,316 2,408
470 471
